Skip to main content
Search

# Context compaction

2 related articles

Subscribe to updates
  1. arXiv

    Agent outputs now trace sentence by sentence: LMLineage tags each source out of band and compacts only documents that drift out of use

    LMLineage maps every output sentence of an LLM agent session to the input spans that support it, with calibrated confidence, without white-box access to the model or extra inference passes. It runs as an out-of-band service, matches by window-level embedding similarity, and uses that lineage to detect which input documents are no longer referenced, compacting only those while keeping user instructions verbatim. On 71 Chinese gold-annotated sentences, the greedy-window path reaches 0.725 mean fine-grained accuracy versus 0.497 for exhaustive line-matching; a separate 90 turns of multi-turn sessions across three generator models serve only latency and feasibility checks.

  2. arXiv

    Long contexts may be unnecessary: agents that compact their working memory keep long-task accuracy while cloud token costs fall several-fold

    On long-horizon and long-input tasks, frontier language models that repeatedly compact their conversation history into a short state, or store long documents as files and read them on demand, keep task accuracy broadly level with keeping the full history, with small gains or losses in individual settings. Because cloud serving charges by context length, shortening the context cuts estimated cost to a fraction. The authors sketch a semantic state space model (SSSM) that carries only a natural-language state between turns.

You’re all caught up